New overkill TEC heat-sink lab host advice

Well, basically I've been playing with the idea of a DIY TEC cooled labby for a few days, and after seeing jayrobs icebox tutorial the idea became more than feasible.

I would like to go one step further and get a custom aluminum lab heatsink milled on a CNC router, so that it's all ready for the different componants I want to include.

- Current reading panel meter
- Adjustable pot to change diode current
- Key switch
- small fan
- TEC cooled heatsink
- Direct press diode heatsink
- adjustable focus
- Unlimited duty cycle

The whole idea of this is so that I can pop in for example a LOC into the removable heatsink and then run it way above the recommended current for extra high outputs whist not sacrificing life span.

I have a proposed design and as long as it's not too expensive to get machined (probably going to be) I'll probably go a head with the project, I just need to sort out the electronics to go in it, such as 3 LM317 based drivers, 1 for the diode and 1 for each TEC.

It would be run of a DC plug from the wall socket, it's just a matter of working out how much power I'm going to need and buying the appropriate plug.

Design:



Fan needs to be moved down slightly


give an idea of scale


The main reason it's pretty large is too allow for enough width at the back to hold the 79mm wide panel meter to read the current.

Advice:

I know this is an overkill, but realistically with this much cooling would it be possible to push a LOC, or a GGW for example much higher than a heavily heats-sinked host? any one got any ideas on numbers?

This project would be really cool if it happens, but Im going to need to find a CNC machinist so if there's anyone out there who thinks they can get 1 or 2 of these made for a good price say so

comments?

If this doesn't work out then I will definitly be giving Jayrobs Ice box tutorial ago with some different components.

You know, the idea is good... BUT I don't think you will be able to achieve AMAZING results with the extra cooling. I mean, the diode should saturate before you can crank a LOC to 1A, if you know what I mean.

Have you considered merging two beams instead of spending the money to get this machined first?

Also, I may be wrong, so I'll let the specialists comment

Don't let this throw you off, man, keep up the good work!!






Also, the game

nikokapo said:

You know, the idea is good... BUT I don't think you will be able to achieve AMAZING results with the extra cooling. I mean, the diode should saturate before you can crank a LOC to 1A, if you know what I mean.

Have you considered merging two beams instead of spending the money to get this machined first?

Also, I may be wrong, so I'll let the specialists comment

Don't let this throw you off, man, keep up the good work!!






Also, the game

Click to expand...

Good idea, I'll definitly look into that as an alternative if this doesn't work out.

I know what you mean about the LOC just refusing too take any more current, or just getting to a point where adding more current simply isn't increasing power which is something I'm worried about.

This could come in handy with things like 12x or GGW's where we haven't really pushed them all the way too there limits and kept them there, I think jay pushed his PHR upto around 190mA which is 60% over the normal current for PHR's.

And you suck! I lost

So where is the TEC in your drawing....:thinking:

I'm always with mindset of keeping Electronics and Heat Sources separated
if at all possible...:undecided:

BTW.... it is very expensive to get a company to make a single prototype on
an CNC milling machine...


Jerry

lasersbee said:

So where is the TEC in your drawing....:thinking:

I'm always with mindset of keeping Electronics and Heat Sources separated
if at all possible...:undecided:

BTW.... it is very expensive to get a company to make a single prototype on
an CNC milling machine...


Jerry

Click to expand...

Yea I thought it might be,

with some luck my school might have a metal drill bit for their CNC machine, as far as I know they only have a wooden one. I was hoping someone outside of a company might own one but that isn't seeming likely.

As for the TEC positioning etc, I'm working on the insides atm, I'll post another model screenie in a bit.

You better read up on the functioning basics of how TECs
actually work... It would not be efficient to mount a TEC in
an enclosed area...

Jerry

I suppose the TEC is inside the big heatsink, and the laser is mounted in some block of sorts onto that tec.

I dont see any use for this elaborate construction with a panel meter and controls mounted in the outer heatsink though. There is no need or use to put them there, it just complicates the design.

If your outer heatsink is big enough, you can put the LM317s on it too, they will not mind operating at higher temperatures, but make sure to electrically isolate them to avoid causing a short.

Hi Ash,

I was trawling a while back and stumbled on this, (rather in depth!), info site regards considerations when using TEC cooling. I hope it helps.

M


P.S. I could've done with a TEC cooled brain whilst reading it!

Morgan said:

Hi Ash,

I was trawling a while back and stumbled on this, (rather in depth!), info site regards considerations when using TEC cooling. I hope it helps.

M


P.S. I could've done with a TEC cooled brain whilst reading it!

Click to expand...

I can't see a link? *confused*

D'oh! .... I'm such a numpty sometimes! (Note to self: Insert copied link before posting quick reply!!!)

Here it is,

The Heatsink Guide - Peltier cooler information

Thanks Adam!

M

Morgan said:

D'oh! .... I'm such a numpty sometimes! (Note to self: Insert copied link before posting quick reply!!!)

Here it is,

The Heatsink Guide - Peltier cooler information

Thanks Adam!

M

Click to expand...

Haha, no problem. We all do it sometimes!

Cheers guys, I'll definitely read up on that link Morgan before I go about planning the interior of this thing.

And yea I do agree the screen isn't really necessary, and in all honesty won't be included that's if I ever actually get this built up.

I'm thinking of just going down the route of Aluminum enclosures from maplins, which should work suitably, unless I can convince my school too buy a metal drill bit for their CNC machine

I think a current readback is usefull because the current is user adjustable. Adding photodiode feedback would make it even more usefull.

I think you can place the fan a bit better, the heatsink fins are on the outside of the case, but the fan blows inside, also taking dust in. Place the fan on top of the fins. Or, if he heatsink is large enough, leave it away. And use some temperature feedback, makes the setup more stable.

Asherz said:

Good idea, I'll definitly look into that as an alternative if this doesn't work out.

I know what you mean about the LOC just refusing too take any more current, or just getting to a point where adding more current simply isn't increasing power which is something I'm worried about.

This could come in handy with things like 12x or GGW's where we haven't really pushed them all the way too there limits and kept them there, I think jay pushed his PHR upto around 190mA which is 60% over the normal current for PHR's.

And you suck! I lost

Click to expand...

An LOC will generally only produce 10-15mW more when pushed from 520mA to 600mA. How-ever heat-sinking does allow you to push them higher with longer life.
I have 7 hours on my 510mA LCC now, a LOC @600mA made it 10 hours.
Believe me, I had a whole lot more then $10 worth of fun in those 10 hours.(45-47 days, used every day)
Both have "overkill" heat-sinks.
In my opinion the only "overkill" heat-sink is one which cools the LD below it's Max. efficiency level.
Here's the latest heat-sink I have been making for the Aixis. It has achieved the best thermal transfer efficiency of all I have made.(of course the little fans play a large role in these numbers@9000 RPM)



Use something like this, (or larger) and build your other peripherals around it. This is all the cooling you need for a LOC, run times of 20 minutes the H/S temp rose 10 degrees. TEC only becomes advantageous at >1W IMHO (cost, space, condensation) not to mention in any TEC the back half of it has to be outside the enclosure. All BTU's of heat removed by one side will be generated on the other side.

^They will all be regenerated, plus a few. It doesn't make any sense to put a TEC in an enclosure most of the time.

Nice little build you have there green, I really like the design.

Since a few have pointed out a TEC cooler would be un neccesery, and simply a large heatsink would be more suitable and a fan maybe the TEC is more hassle than it's worth.

I think moving the fan too the top would make more sense, but just too make it clear that would be sucking air out of the enclosure not blowing in, I think I want too keep the small screen in the design, it will make it a lot easier to set current etc if I want to swtich diodes around.

I've been searching around for some local CNC machinists without much luck so far, if anyone knows of anyone willing to do some one off products on the CNC please let me know.

I've been trying too work on some interior plans in between school work and have something going, I might post some sketches, since I don't really enjoy using CAD much and finding sketching easier and quicker.

Thanks for the ideas and help guys